© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19554
    Prev
    Next
    INDEX
    Explanations

    declare @sql

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åīįæıIJæĺ¯
    -0.09
    æľĢåĪĿ
    -0.08
    ä¸ĵåįĸ
    -0.08
    è¾½
    -0.08
    alte
    -0.07
    z
    -0.07
    http
    -0.07
     пеÑĢвого
    -0.07
    广éĺĶçļĦ
    -0.07
     Fal
    -0.07
    POSITIVE LOGITS
    ä¹Łæĺ¯éĿŀ常
    0.10
    ä¸Ģå¦ĤæĹ¢å¾Ģ
    0.09
    ä¹ŁåºĶ该
    0.09
     GLsizei
    0.09
    éĥ½èĥ½å¤Ł
    0.09
    âľª
    0.09
    åĢ¼å¾Ĺåħ³æ³¨
    0.09
    åįĵ
    0.09
    ategy
    0.09
    ä¸įåłª
    0.09
    Activations Density 1.307%

    No Known Activations